Search

In April 2009 two pediatric cases of Influenza-Like-Illness (ILI) were reported in southern California and were positive for swine influenza A (H1N1) virus (MMWR, 2009). Heightened surveillance is needed to track and monitor this emerging pathogen. One of the main ideas to strengthen capacity was heightened surveillance or syndromic surveillance. Syndromic surveillance is the systemic and ongoing assessment of the syndromes, including the timely collection, analysis, interpretation, dissemination, and subsequent use of data. Surveillance of Over-The-Counter (OTC) medication sales and Emergency Department (ED) syndromic surveillance are just a few of the new methods developed for early detection of outbreaks. These new surveillance techniques utilize computer databases and statistical packages to store and analyze the data. New/seasonal outbreaks in Montgomery County in 2007-2010 were used to retrospectively interpret data. These outbreaks included novel H1N1, seasonal influenza, and seasonal norovirus that occurred in Montgomery County, Maryland. This study used data from the Electronic Surveillance System for Early Notification of Community-based Epidemics (ESSENCE) syndromic surveillance system. Several syndromic categories were used and created to retrospectively determine at what point in time the outbreak would have been detectable. The categories include: Seasonal/H1N1 ILI, and Gastrointestinal Illness. Regression/Exponentially Weighted Moving Average (EWMA) 1.1 was used to analyze the Montgomery County Health Department's syndromic surveillance data. This research explored whether ED or OTC data provide an alert first and what is the temporal difference between the two. The hypotheses tested for this study was: 1) There is a temporal difference between the purchase of OTC medications and ED visits in a given community with ED visits alerts posting before OTC medications among Seasonal flu and H1N1 outbreaks. 2) Individual hospitals ED's with the smallest patient loads will have a different pattern of alerts than the larger ED's. This study found that an alert will not always appear in OTC data before it appears in ED data depending on the type of outbreak and alert criteria used by the epidemiologist. The alert criteria are crucial in determining when an outbreak is suspected and when to initiate a response from public health. Large hospitals mirrored countywide data. Smaller hospital's ED visits are too sparse to be representative of countywide trends. There is no way to determine whether ED data or OTC data will produce an earlier signal. The different symptomatology and type of outbreak will affect the timeliness of alerts from each data stream. The findings of this study will benefit local and state government agencies response to an outbreak/bioterrorist event. Understanding how OTC sales and ED data interact can lead to an early response that can limit the spread of an outbreak. The knowledge gained will enhance the response plans developed by health agencies to better address the unique needs of populations in Maryland.